Informal caregivers consistently report higher rates of stress than non-caregivers, and there is evidence that chronic stress has adverse effects on physical health and cognitive functioning. However, no study has evaluated the biological processes between caregiving-related stress and health decline. Recent theories of stress-induced metabolic syndrome and health decline suggest a plausible mechanism: chronic stress disrupts several interrelated homeostatic systems (neuroendocrine, HPA axis, and immune systems), which lead to metabolic syndrome and subsequent health decline. The proposed study will evaluate these associations over a 3-year period in 400 elderly caregivers and non-caregivers from the Boston, MA metropolitan area. Caregivers will be recruited from the Boston University Alzheimer's Disease Core Center Registry (n=200). Non-caregivers will be recruited from the Harvard Cooperative Program on Aging Registry (n=200). Data will be collected during 4 annual interviews at the BU General Clinical Research Center. Biological measures will be obtained from 24-hour urine (catecholamines and cortisol), serum (DHEAS, IL-6, glucose, triglycerides and HDL cholesterol);blood pressure and anthropomorphic measures will be taken. Data on stressors, the caregiving situation (measures from Pearlin's Caregiver Stress model), and potential confounders and effect modifiers will be collected through interviewer-administered questionnaires at each interview. Functional status will be based on walking speed, chair stands and self-reported ADLs. Cognitive status will be based on multiple domains (e.g., executive functioning, working memory). Hypotheses will address: 1) the sequential pathway from stress, to physiological changes, metabolic syndrome, and functional and cognitive decline;2) comparisons of these outcomes in non-caregivers and non-caregivers;and 3) the mechanisms by which modifiers, such as social support and caregiving characteristics, affect this pathway. Mixed effects models will be used to analyze the longitudinal process between caregiver stress and functional and cognitive changes. Latent growth curve modeling will be used to examine the effects of time-varying covariates (i.e., caregiver transitions) on intra-individual change across the four measurement points. Results will provide insight into biological pathways underlying health decline in aged caregivers, and have implications for interventions to reduce adverse effects of chronic stress.